Artificial Intelligence-Based Cardiac Rehabilitation Therapy Exercise Recommendation System

Cardiovascular disease is one of the leading causes of death in modern society. There are many factors that must be taken into consideration when analyzing cardiovascular disease (CVD), such as a patient's quality of life, general well-being, diet, lifestyle, psychosocial effects, medical costs, and many other factors. Pharmaceutical interventions are the primary method of treating heart conditions, however, one of the most popular and widespread methods of treatment which often augments medication is Cardiac Rehabilitation Therapy (CRT). CRT programs have been shown to be an effective secondary prevention method for CVD. It can help reduce patient risk, as well as improve their recovery and overall health outcomes. In this paper, we propose a recommendation system for patient decision support, and a computational model for cardiac rehabilitation planning, in order to personalize and optimize the physical exercise therapy plan offered to patients, to meet their goals and needs. We tested our proposed model against randomly generated exercise plans, and found that our model was able to find the optimal therapy plan with a 43% higher rate of success on average relative to the random model, with regards to meeting patients' goals and needs. We believe our patient-centered solution to CRT will improve patient satisfaction, quality of life, and help reduce costs through improved efficacy and adherence to the therapy program, along with improved post-therapy cardiac health management.